HF Model & Benchmark Matrix

24,078 variants · 14,984 scores · 120 benchmarks

Breeze-ASR-26

MediaTek-Research/Breeze-ASR-26

kind
original
quant
bf16 (16-bit)
family
misc
modality
automatic-speech-recognition
root model
openai/whisper-large-v2
provider
MediaTek-Research
license
apache-2.0
language
nan, zh
context
—
library
transformers
vRAM (GB)
—
size on disk
—
downloads
5,483
likes
69
created
—
last modified
2026-04-21T07:00:40+00:00 (2026)
huggingface
https://huggingface.co/MediaTek-Research/Breeze-ASR-26
tags
transformerssafetensorswhisperautomatic-speech-recognitiontaiwanese-hokkientaigilow-resource-languagefine-tunednanzharxiv:2603.19259base_model:openai/whisper-large-v2base_model:finetune:openai/whisper-large-v2license:apache-2.0endpoints_compatibleregion:us

Benchmark scores (1)

benchmarkcategoryscorecontextconfevidence
CERspeech-asr30.7low| Taiwanese Input Method (教育部台灣台語輸入法) | 30.70 |

Discussions (0)

none scraped.

Hugging Face model & benchmark data scraped & matrix-ified. 24,078 variants · 14,984 scores · 120 benchmarks.